Title :
Sentiment classification for online comments on Chinese news
Author :
Fan, Wen ; Sun, Shutao
Author_Institution :
Sch. of Comput. Sci., Commun. Univ. of China, Beijing, China
Abstract :
With the development of network technology, all kinds of events will be shown up as news rapidly on the World Wide Web. Internet users read the news and some of them give their comments online. It is important for crisis public relations, government decision-making and news impact analysis to understanding the netizens´ comments on the news events. Due to the huge amount of news and comments on Web, it´s very difficult to collect and process them manually. This paper proposes the framework of a sentiment classification system for online comments on Chinese news, and discusses the implementation technologies of news comment collecting and classifying. Experimental research is also conducted, and the result shows that the Support Vector Machine (SVM) approach usually achieves better performance than k-nearest neighbor (kNN) approach.
Keywords :
Internet; decision making; information resources; pattern classification; psychology; support vector machines; Chinese news; Internet users; SVM; government decision making; network technology; news impact analysis; online comment; sentiment classification; support vector machine; world wide Web; Accuracy; comments; kNN; sentiment classification; support vector machine;
Conference_Titel :
Computer Application and System Modeling (ICCASM), 2010 International Conference on
Conference_Location :
Taiyuan
Print_ISBN :
978-1-4244-7235-2
Electronic_ISBN :
978-1-4244-7237-6
DOI :
10.1109/ICCASM.2010.5619288
